Re: [Sipping] Testing the water: A transaction torture-tests draft

Hi Byron,

this sounds like a useful document. I think it will be appreciated by 
implementers. I personally like (very much) specs that document 
implementation experiences and that make the life of future implementers 
easier.

Thanks,

Gonzalo

Byron Campen wrote:
>     This is mostly Dialog-related weirdness (although there are a few 
> transaction-level cases here). Most of the examples are things that can 
> arise accidentally between compliant endpoints. I was thinking a draft 
> that would focus exclusively on transaction-level cases, including cases 
> where there is a broken/malicious endpoint sending complete garbage. 
> (For instance, ACK in a NIT, response to an ACK, sending provisionals 
> after a final response, mixed final responses, an INVITE collides with a 
> previous ACK/200's transaction id, etc)
> 
> Best regards,
> Byron Campen
> 
>> Hi Byron,
>>
>> are you thinking of something similar to the draft below?... or 
>> something else?
>>
>> http://www.ietf.org/internet-drafts/draft-ietf-sipping-race-examples-01.txt 
>>
>>
>> Cheers,
>>
>> Gonzalo
>>
>>
>> Byron Campen wrote:
>>>     I was thinking a torture-tests draft in the same spirit as RFC 
>>> 4475 and draft-ietf-sipping-ipv6-torture-tests; an enumeration of 
>>> test-cases that could cause trouble for an unsuspecting 
>>> implementation, with exposition on what makes the case interesting, 
>>> and how implementations should react. Of course, the tests themselves 
>>> would read differently, since the contents of the messages won't 
>>> matter apart from stuff like the method and branch params.
>>> Best regards,
>>> Byron Campen
>>>>    From: Byron Campen <bcampen@estacado.net>
>>>>
>>>>    I have doing a lot of work lately on hardening a SIP stack, and I
>>>>    have repeatedly come across bugs triggered not by syntactic garbage
>>>>    (malformed header-field-values and the like), but by semantic
>>>>    garbage, usually in the form of transaction collisions. I'm trying
>>>>    to get a feel for working-group interest on the subject, so let me
>>>>    know what you think.
>>>>
>>>> It's obviously very valuable work.  What are you proposing?
>>>>
>>>> Dale
>>>>
>>>>
>>>> _______________________________________________
>>>> Sipping mailing list  https://www1.ietf.org/mailman/listinfo/sipping
>>>> This list is for NEW development of the application of SIP
>>>> Use sip-implementors@cs.columbia.edu for questions on current sip
>>>> Use sip@ietf.org for new developments of core SIP
>>> ------------------------------------------------------------------------
>>> _______________________________________________
>>> Sipping mailing list  https://www1.ietf.org/mailman/listinfo/sipping
>>> This list is for NEW development of the application of SIP
>>> Use sip-implementors@cs.columbia.edu for questions on current sip
>>> Use sip@ietf.org for new developments of core SIP
>>
> 

_______________________________________________
Sipping mailing list  https://www1.ietf.org/mailman/listinfo/sipping
This list is for NEW development of the application of SIP
Use sip-implementors@cs.columbia.edu for questions on current sip
Use sip@ietf.org for new developments of core SIP